Design of Dynamic Traffic Grooming Algorithm in Software-Defined Wireless Mesh Networks

Abstract

Software-Defined Networking (SDN) separates the network control plane from the packet forwarding plane, and provides the upper applications with an global view of the distributed networks state. By making the network device software programmable, this newly emerging networking paradigm can contribute to change the large and highly complex situation of current networks protocol architecture. SDN enabled Wireless Mesh Networks (WMNs) will apply SDN technology into WMNs. By separating network control logic from data forwarding, automatic control and update functions would be achieved by software-driven control logic, which improves network performance and shortens the cycle of network innovation. Traffic engineering problem involves in the task of mapping network traffic distribution to existing physical topology. It can control network resources utilization and fair traffic allocation, which will optimize the utilization of network resources, and solve the imbalance problem of network resources, so that the network will reach a high degree of reliability, robustness and operability. Therefore, the design of dynamic traffic grooming algorithm in SDN enabled Wireless Mesh Networks is of important theoretical significance and practical value. This article focuses on traffic engineering technology in software defined networking. Aiming at the dynamic routing problem, the SDN controller dynamic traffic grooming algorithm is designed, which takes into account both the path length and link utilization to determine routing establishment, so as to achieve the network load balance and reduce the network congestion. Simulation results demonstrate that our proposed dynamic traffic grooming algorithm performs better in terms of average end-to-end throughput, average end-to-end delay and packet loss probability than traditional scheme.